Halsey H. Matteson

Effects of Surround Size and Luminance on Metacontrast

The temporal course of reduction of test-stimulus effectiveness under conditions of metacontrast was studied with a suprathreshold brightness-matching procedure in which test-stimulus luminance was varied to maintain constant brightness. One purpose of this study was to determine the effects of surround diameter on the over-all amount of reduction of test-stimulus effectiveness. Two specific predictions concerning the effect of surround size on the temporal course of reduction of test-stimulus effectiveness were based on the notion of delays introduced in neural pathways mediating lateral inhibition. The typical metacontrast effect was obtained, i.e., maximum reduction of test-stimulus effectiveness occurred when the test stimulus preceded the surround by an appreciable interval. The amount of metacontrast increased with increasing surround diameter, but surround size had no consistent effects on the temporal course of reduction of test-stimulus effectiveness. Although the amount of metacontrast decreased with decreasing surround luminance, the stimulus-onset asynchrony at which maximum reduction occurred showed no consistent changes with decreasing surround luminance.

The range of spatial frequency contingent color aftereffects.

Abstract Checkerboard stimuli contain two-dimensional Fourier components oriented 45 degrees from edges of the individual checks. Adaptation to such stimuli and testing with rectilinear gratings showed that contingent color aftereffects are associated with the Fourier components rather than the edges of such patterned stimuli at high spatial frequencies. With low spatial frequency (0.8 c/d) contingent color aftereffects were aligned with the edges rather than the Fourier components. These data, obtained with a color cancellation procedure, are in agreement with previous reports of spatial frequency-contingent color aftereffects and indicate the range over which such effects can be obtained.

Effects of Surround Luminance on Perceptual Latency in the Fovea

The effect of surround luminance on relative latency of response to a test stimulus was measured with the perceived-order method in which asynchrony between two lights is varied to make the lights appear subjectively simultaneous. Increasing surround luminance from zero to levels high enough to impair test-stimulus detectability resulted in reduction of test-stimulus latency (facilitation of response speed) by 100 ms both in the rod-free area of the fovea and in the periphery. Since appreciable facilitation of response speed was obtained in the rod-free area of the fovea, suppression of visual noise would seem to be a more plausible explanation of the facilitation effect than inhibition of rod activity. Variation of surround luminance was also found to have a much greater effect than variation of test-stimulus luminance.

Effects of Surround Luminance on Perceptual Latency

The apparent brightness of a test field can be substantially reduced by an extended-duration surround. In experiment I, test-field luminance was adjusted to match the brightness of a reference stimulus. Over a wide range of surround luminance, latency was measured by comparing test-stimulus and surround-field onsets, using the perceived-order method, in which the temporal interval between the two stimuli is varied to obtain subjective simultaneity. A series of control observations was made using test stimuli covering the same range of test-stimulus luminance that was used in the main series of observations, but with no surround field. Latencies of test stimuli were found to be considerably shorter with high-luminance surround fields than with no surround. Apparently, high-luminance surrounds resulted in reduction of latency of response to the test stimulus (facilitation of response to the test stimulus). In experiment II, a constant test-stimulus luminance was used. Reduction of latency of response to the test stimulus was found to increase with increasing surround luminance when surround luminance exceeded test-stimulus luminance. Supplementary observations indicated that the reduction of test-stimulus latency was not an artifact of dark adaptation.

Interacting spectral sensitivity functions obtained in a contrast situation

Abstract The spectral sensitivities of the interacting processes which mediate brightness contrast have been measured at low luminance levels. These functions show that rods may be inhibited by cones. The data are discussed in relation to the results of Alpern and Rushton, who concluded that lateral interactions occur only between processes derived from the same type of receptors. Differences in receptor latencies are suggested as a possible mechanism for obtaining some of the advantages derived from a specificity of lateral interactions, even if the spatial interconnections are non-specific.

The effects of differential adaptation on spatial frequency-contingent color aftereffects

The magnitude of color aftereffects was measured with a color-cancellation technique after subjects adapted to oblique checkerboards in magenta light and square-wave gratings of various orientations in green light. Test stimuli were composed of achromatic patterns of an oblique checkerboard and an appropriately oriented grating. Two different spatial frequencies were used: 0.8 and 3.0 c/deg. With 3.0-c/deg stimuli, the largest color aftereffects occurred when fundamental Fourier components of adapting stimuli were oriented at 45° angles to one another. At the lower spatial frequency (0.8c/deg), the greatest color aftereffects occurred when fundamental Fourier components of adapting stimuli were in the same orientation. These findings agreed with previous color aftereffect studies which involved adapting to checkerboards and testing with gratings. They offer further support for the notion that the human visual system responds to spatial frequency content of such stimuli as it is described by two-dimensional Fourier analysis.

Perceptual latency as a function of stimulus onset and offset and retinal location

Abstract Relative perceptual latency to stimulus onset and to stimulus offset was determined in the fovea and parafovea using the perceived-order method. Offset latency was shorter than onset latency at all luminance levels in the fovea. In the parafovea, however, offset latency was found to be shorter than onset latency only at low test stimulus luminances; offset and onset functions converged at higher luminance levels. Contradictory findings obtained in early phases of the present study and conflicting results existing throughout onset-offset literature are discussed in terms of visual persistence.

Spatio-temporal processing in multiple sclerosis

The processing of spatial and temporal detail was investigated in patients with multiple sclerosis. Normal observers and 13 patients with optic neuritis secondary to multiple sclerosis performed a battery of visual tests that included contrast sensitivity, temporal integration, evoked potentials, and visual masking. The multiple sclerosis patients exhibited losses of pattern processing, and these deficits became more noticeable when the patterns were presented briefly. Moreover, these patients exhibited diverse response patterns for the different visual tests. For some, temporal integration functions appeared severely attenuated, while evoked potential latency was within normal limits. Others displayed poor performance in the visual masking test, yet contrast sensitivity functions were comparable to those of the control group. We suggest that a battery of tests that incorporates spatial as well as temporal stimuli is necessary for the detection of visual dysfunction in multiple sclerosis.

Metacontrast in the fovea

Metacontrast is a special case of backward masking in which maximum masking occurs when test precedes mask. Breitmeyer and Ganz (1976) proposed that metacontrast is due to inhibition of sustained test rksponse by transient mask activity, and a task requiring resolution of high spatial frequencies should result in maximum masking when test precedes mask. Breitmeyer et al. (1974, 1976) and Breitmeyer (1978b) used a task requiring judgment of presence or absence of small straight segments (chops) on the edges of two disks, and masking was greatest when test preceded mask by about 75 msec. Alpem (1953) found more metacontrast with increasing distance from the fovea, and he found no metacontrast in the fovea. Breitmeyer and Ganz (1976) stated that metacontrast should be weak or absent in the fovea, since Fukuda and Stone (1974) found that transient cells were infrequent in the center of the cat’s retina. However, Fukuda and Stone did find transient cells in the center of the retina. With narrow mask and test, Saunders (1977) found a 0.7 log metacontrast effect in the fovea, which was only slightly smaller than the effect in the parafovea. Saunders found no effect with large test and mask in the fovea and argued that earlier failures to obtain fovea1 metacontrast might be due to large stimuli. Metacontrast has been obtained in the fovea with small stimuli and brightness ratings (Bridgeman and L&T, 1980; Stoper “do Banffy, 1977), vernier acuity thresholds (Westheimer and Hauske, 1975). a percent detection measure of vernier acuity (Breitmeyer, 1978a), threshold for deviation from vertical (Westheimer et al., 1976) and stereoscopic acuity (Butler and Westheimer, 1978). In the present study, a single chopped-off disk was rotated to one of four orientations. This task makes it possible to confine stimuli to a smaller area than Breitmeyer’s procedure. The purpose of this study was to determine if metacontrast (maximum masking when test precedes mask) could be obtained in the fdvea with this task and to compare fovea1 and parafovea1 metacontrast functions. Since all studies with Breitmeyer’s chopped-off disk task had no conditions in which mask preceded test, another purpose was to determine the extent of forward masking with this task.

Comparison of Two Measures of Metacontrast

The temporal course of reduction of test-stimulus effectiveness under conditions of metacontrast was studied with two different suprathreshold brightness-matching procedures. (1) Test-stimulus luminance was varied to make apparent brightness of the test stimulus equal to the brightness of a constant-luminance reference stimulus. (2) Reference-stimulus luminance was varied to match the apparent brightness of a test stimulus maintained at constant luminance. Functions obtained from two observers when the test-stimulus luminance was varied were typical metacontrast functions with maximum reduction of test-stimulus effectiveness when the test stimulus preceded the background by 50 ms, and the shape of the functions was highly similar to functions obtained previously with a disk–ring stimulus configuration. Functions obtained by varying reference-stimulus luminance were highly variable both between observers and within observers. The difficulties encountered with this second method probably resulted from glare from high-luminance test and surround fields.